WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Screen Readers as a Development Tool for Web Developers

for

From: Jennifer Sutton
Date: Jul 17, 2015 3:46PM

Paul and all:

Since Dave and I are lifelong screen reader users (for as long as
they've existed, I believe), and we have decades of experience in
this industry (between us), we're well aware of this fact -- that
ARIA, as it works today, requires a screen reader to test it.

But my point is that the current approach ain't workin' in the real world.

Sure, it works for businesses who are in the business of
specializing. And you bet I speak for myself, as a self-employed part
of this industry. Just because I can directly benefit from the status
quo doesn't mean I think it's the right path.

Today's expectation that devs/designers/whomever
(someone?/anyone?) should test with a screen reader isn't working out
very well for end-users since screen reader testing sites that
implement ARIA (no matter who does it) is not scaling. I don't see
that the notion is realistic. ARIA should *not* require screen reader
testing since it may be (today), but doesn't *have* to be, only for
screen reader users.

People who work *for* blind people in this industry, but who aren't
blind themselves, typically can go back to "normal" surfing for their
everyday tasks/needs, so the constant presentation of "ARIA going
even wilder, and wilder, and wilder . . ." (and flat out often being
a "blocker" when implemented incorrectly) is less for many of you
than for others of us. It's EVERYWHERE! And it's so rarely
implemented correctly that I daresay end-users have few opportunities
to learn how it should work.

[Thanks for the theft and modification of one of your presentation
titles, Jared]

As I was trying to do, in the previous thread, I'm not thinking here
about today; I'm trying to think about a different future, where we
learn from what isn't working in the present, and we maximize what
ARIA could be bringing to the table (and perhaps accessibility in
SVG, too, for that matter).

It's pretty simple, in my mind. Generally speaking, if sighted people
can't see it, they're not going to do it. By "seeing it" I mean
integrating prompts in tools, or some kind of screen reader emulator
that visually shows standard screen reader behaviors. Or something
even cooler that I can't imagine.

And even if those who are working on sites *could* see it, they'd
still have to go the extra mile to understand what they're seeing,
but at least the learning curve would be less.

That's just human nature; I'm not bashing anyone at all, here. I keep
thinking that if blind people expect sighted people to accommodate
us, then we need to figure out a way to accommodate them, even as I
wish this were about more than blind folks.

I'm so very excited to see tools, like Tenon, that are focusing on
being incorporated into common dev tools, to meet people where they
are. And, though I've not tried them, I'm seeing some mobile teaching
and testing tools that may have this idea in mind, to a degree.

Clearly, I'm thinking "bigger" than today's approach.
We've had many years of history in this industry (with progress,
but progress that's pretty slow in terms of adoption); it'd be great
if we could shift from repeating it, to find some ways to speed things up.

With apologies, Dennis, for side-tracking the conversation.

Jennifer, who will now go back to work in order to test and fix what
she can and stop beating this poor horse